Good day and welcome to the Publicis Third Quarter 2020 Revenue Conference Call. Today's conference is being recorded. At this time, I would like to turn the conference over to Mr. Arthur Sadoun. Please go ahead, sir.
Merci, Cécilia. Bonjour, and welcome to Publicis Groupe Third Quarter 2020 Revenue Call. I am Arthur Sadoun, and I'm here in Paris with our CFO, Jean-Michel Etienne, and our Secretary General, Anne-Gabrielle Heilbronner. Steve King, COO of Publicis Groupe, is joining us from London. As usual, the directors will answer all of your questions together after the presentation. Alessandra Girolami is also in Paris and will be available to take all of your questions offline after this session, which will end at 10:30 Paris time. Please take the time to read the disclaimer, which is an important legal matter. Now, let's dive into the presentation. If I may, I want to start by giving some context to the current global situation. While the virus and its economic and social effects continue to spread, market conditions have improved, as anticipated.
According to forecasters, global ad spend in Q3 is broadly at -10% after Q2 at -23%. When it comes to clients, they have continued to shift their spend to meet the changing consumer behaviors that have emerged in the last six months. We have seen them expand further into digital channels and increase their brand investment in e-commerce and direct-to-consumer. Those trends will continue in the post-COVID era. Looking at the rest of the year, I can tell you, after meeting many clients, that their fighting spirit is strong despite the uncertainty ahead.
They are ready for life with COVID. This means reinventing their business model when their industry has stumbled, accelerating their strategy to come back to growth in sectors that have been less impacted, and maintaining market share in categories that have been able to expand during this crisis.
In this context, and thanks to our unique model, Publicis has been less affected than the market. We have been able to capture part of the shift of our client investment. This has gone some way to mitigating the revenue decline we have seen due to the crisis. In Q3, our organic growth is at -5.6%. After a good start of the year and a resilient Q2, our nine-month organic growth is at -7.2%. These results demonstrate that we have the right assets and strong financial backbone to weather this crisis. Most importantly, we have our people to continue to manage the situation extremely well, both personally and professionally, while most of them are still working from home. I want to thank them deeply for their dedication. There are four highlights of our Q3 numbers. First, the US.
We delivered significant sequential improvements in Q3 with a -2.4% organic growth in a market that now represents 60% of our business. Taking into account our resilient performance in the first six months, we posted -3% growth on a year-to-date basis. After outperforming the industry in Q2, Q3 growth in the U.S. was again ahead of our industry estimates at -7% for advertising expenditure. That performance was supported by the solid revenue streams we built in the last quarters thanks to both new business but also cross-fertilization.
We have experienced double-digit growth on some strategic areas, including our health solution, our production activity, and our programmatic digital media arm. After some consulting projects were put on hold in Q2, Publicis Sapient U.S. showed a resilient performance and posted organic growth in line with our U.S. operations. Looking ahead, the pipeline is encouraging for the coming quarters.
Finally, Epsilon played a vital role in this performance as a key differentiator for our overall offering. It was included in our organic growth for the first time in Q3. With a strong improvement in Q3 versus Q2, it positively contributed to our US organic growth by being flat. Second highlight, Europe posted significant sequential improvements in Q3 versus Q2. As advertisers resumed spending after lockdown in their region, organic growth was at -9% in Q3 compared to -23.5% in Q2. We saw these trends almost in every country. The U.K. was down at -10.6% organically, halving its rate of decline thanks to improved performance in media and creative. A similar pattern was visible in France with organic growth at -13.8%. Our outdoor media operation reopened. However, their activity remained impacted by slow resumption of the public transport use.
These activities had a negative impact of 400 basis points on France's organic growth in Q3. Germany posted negative organic growth at -8.6%. Italy and Spain were positive this quarter with organic growth at +7.8% and +8.6% respectively with new business momentum. Let's turn to our third highlight. In Asia, organic growth was at -9.2% with several countries still impacted by the crisis. Greater China's organic growth remained negative, mainly reflecting its exposure to some industries still impacted by the crisis. Finally, we continue to demonstrate the competitive nature of our model with significant wins, like Kraft Heinz, Reckitt Benckiser, TikTok Global Media, Global Production at GSK, and BT Creative and Production Accounts, among others. As you can see, we do have strong foundations to weather this crisis thanks to our uniquely integrated model.
This is why we are posting organic growth that is above average industry estimates. That resilience was also visible financially, as demonstrated by our decision to fully refund our RCF in September, backed by our solid balance sheet. I will now hand over to Jean-Michel, who will take you through the detail of our organic numbers and other specific financial indicators.
Thank you, Arthur. Good morning, everybody. I hope that you and your families are still well and safe in this very specific context. We now detail our third quarter and first nine months 2020 regarding net revenue and net debt. I will then give the floor back to Arthur, who will go through our priorities and the outlook for the remaining part of the year. Let's start with net revenue on page eight. In Q3 2020, net revenue is EUR 2,343,000 , down 9.1% versus last year as reported. Currencies have a 3.7% negative impact with the US dollar, explaining half of the impact. Acquisitions have no impact at 0%, given that Epsilon is now included in our organic performance for the first time this quarter. All in all, organic growth is at minus 5.6% in Q3.
For the first nine months of 2020, net revenue is EUR 7,117,000 , up 2.7% versus last year as reported. Currencies have a 0.9% negative impact. Acquisitions have a positive impact of 10.7% due to the consolidation of Epsilon. Therefore, the organic growth for the first nine months of 2020 is at -7.2%. On page nine, I will be quick as I will detail the performance by geography in the following slides. Organic growth in Europe is at -9%, significantly improving versus Q2, or -8%, excluding the impact of our French outdoor activity and the Drugstore . North America is delivering an organic growth of -3%, a significant improvement versus Q2, and a solid performance in the current context. In Asia-Pacific, organic growth is -9.2%, impacted by a high comparable basis when compared to Q3 2019. Latin America is at -14.8%.
Middle East and Africa are at -11%. Overall, Q3 organic growth for the group is at -5.6%. On page 10, I will detail the performance of our main countries in Europe. First, in the U.K., which is our biggest country in Europe with 8% of our net revenue. Organic growth is at -10.6%. Our creative and media activity is broadly in line with the growth of the country, with media slightly ahead and creative not very far from the average. Publicis Sapient is declining by mid-teens due to the phasing of the project-based activity. Regarding Epsilon 2.0, it is growing fast, although on a small base, as they start their international expansion from the U.K. Our U.K. health business now is also delivering strong growth in the quarter.
In France, which represents 6% of our net revenue, organic growth is -13.8% in Q3, or -9.8%, excluding the impact of our outdoor media activity and the Drugstore . Media is slightly positive. Regarding our creative business, which represents more than half of our activity in the country, it starts to recover from Q2 but remains negative. The performance in France is also impacted by the phasing of some projects at Publicis Sapient France. Finally, Germany, 3% of the group net revenue, is posting an organic growth of -8.6%, supported by the positive contribution of our creative activities. This is thanks to the new business wins. Media is in line with the growth of the country, and Publicis Sapient, which represents less than 15% of the country, is down significantly during the quarter.
On page 11, I will only comment on the performance of the US, which is our biggest country, accounting for 60% of the group, as Arthur said, of our group net revenue. We delivered in the US a solid performance with an organic growth of -2.4% in Q3, demonstrating the strength of our model. Our creative activities, which represent roughly 25% of the country, slightly deteriorated in Q3 versus Q2, partly due to the phasing of some clients' incentives, and it was expected. Our media activities, accounting for nearly 30% of the country, are broadly in line with the average growth of the US. Please note also that we delivered a strong growth in digital media investments.
Epsilon 2.0, which is the core of Epsilon and accounts for 20% of the US, is flat in Q3 and recovering versus Q2, notably thanks to the improvement of the automotive practice and Conversant. Publicis Sapient, around 15% of the country, is broadly in line with the growth of the US, which is, of course, very encouraging. Last but not least, Publicis Health US further accelerates its growth in Q3, posting a double-digit growth rate. On page 12, we detail three regions that account together for 15% of the group net revenue. First, regarding Asia-Pacific, the region is impacted by a high comparable basis versus Q3 2019, especially in China. Growth is slightly positive for our media activities, supported by China and India. In the meantime, the performance of our creative activities is more contrasted, mostly in Southeast Asia.
Publicis Sapient records a strong growth in the region, mostly in India and Australia. Regarding Latin America, the decline in organic growth is mostly due to Brazil, which is, as you know, still heavily impacted by COVID. Middle East and Africa are down by 11%, significantly improving sequentially versus Q2, with the balanced performance depending on countries. On page 13, regarding the first nine months net revenue by geography, I will be quick. Europe is at -14.1%. Growth would have been -11.4%, excluding our French outdoor media activities and the Drugstore . North America is at -3.4%. Asia-Pacific is at -5.9%. Latin America is at -15.4%. Middle East and Africa are at -11.5%. Overall, Publicis Groupe organic growth for the first nine months is at -7.2%.
On page 14, as usual, I will give you some color on our organic growth performance in some countries. I will mention only a few of them. In positive territory, we have Italy at +7.8%, Spain at +8.6%, and New Zealand at +15%. Between 0% and -10%, Australia is at -1.3%, and the Netherlands are at -5.3%. Between -10% and -20%, Brazil is at -17.2%, and Canada is at -13.8%. In the current context, some countries continue to be highly impacted by the pandemic. This is still the case of Mexico and South Africa, which are at around -26%. Now, moving to net financial debt on page 15. Average net debt at the end of September is EUR 3,584,000 , up by nearly EUR 2 billion year on year.
This is only due to the full impact of Epsilon acquisition and the calculation of the average net debt during the first nine months versus only three months at the end of September 2019. Net debt position at the end of September 2020 is at EUR 3,181,000 , a significant reduction versus last year thanks to a strong focus on deleveraging and day-to-day cash management in the context of the crisis. To finish my presentation on page 16, as you can see, our liquidity position has been relatively stable since the end of June. You notice also that we fully refunded the remaining part of our EUR 2 billion revolving credit facility that we preventively drew down in Q1 to face any potential liquidity impact relating to the pandemic.
We first decided to refund half of the line in June, and then we decided to fully refund the second half in September, as we were reassured by our Treasury forecast. I will now hand back to Arthur, who will give you an update on our strategic outlook, and I remain available for questions during the Q&A session.
Merci, Jean-Michel. As you can see, our performance has been ahead of the industry at this stage, both for organic growth and our financials.
This is thanks to the work we have done to build and roll out our offer. By putting data and technology at the core of our creative and media operation, we have what our clients need to unlock growth while reducing costs in this life with COVID and beyond. Data platform to build, enrich, and connect our clients' data assets with our first-party data.
This is the only way to be truly customer-first in a cookieless world. Personalized content that can deliver groundbreaking creativities adapted to real-time customer experiences. A leadership position in media in the US to activate this personalized content at scale and optimize reach wherever the channel. Last but not least, technology to create direct channels with customers in a marketplace where our clients are constantly challenged by direct-to-consumer brands. The integration of Publicis Sapient and Epsilon with our creative and media operation through the Power of One model gives us a unique competitive advantage.
This is visible in our new business track record over the last 18 months and once again in Q3, but also with the performance of our top 200 clients. As you can see, despite automotive and financial services that have been negatively impacted, we are actually posting positive growth of 0.8% for our top 200 clients.
Healthcare is an important driver, but it is also interesting to note that FMCG is at +5.5%. Not only have we built the right offer over the last years, but we have also brought three structural competitive advantages to our organizations that make us confident we will continue to outperform the industry financially in the long term. Our global delivery centers, our country model, and Marcel will enable us to deliver our cost reduction plan of $500 million this year while preserving our talents, rewarding our teams, and investing in the future. Our performance so far this year demonstrated that we continued to weather the crisis pretty well, as we are posting organic growth above industry average estimates for the second quarter in a row thanks to our resilient model.
When it comes to our short-term outlook, we believe that these relatively solid business trends will continue until the end. However, with the acceleration of the pandemic and the restrictions imposed, we have to be cautious about Q4, which might be impacted further and come below Q3. In this context, our cost management makes us confident in delivering our cost reduction plan this year and an operating margin rate slightly ahead of average analyst consensus at 14.3%. To conclude, these are unprecedented times for everyone, and this is unlikely to change soon. We don't underestimate the challenges in front of us all, but we are convinced that we are well equipped to face them.
We have a proven model that is driving value and demonstrating its ability to weather the crisis. We are delivering what our clients need by seamlessly connecting data, creative, media, and technology.
We have our unique structure where we are well placed to continue to do better than the industry financially. Now, if I may, I would like to once again thank our people for their hard work and dedication in those tough times, but also our clients for their trust and their partnership. Thank you for listening, and we'll now take all of your questions.
Thank you. If you wish to ask a question at this time, please press star one on your telephone keypad. Please ensure the mute function on your telephone is switched off to allow your signal to reach our equipment. Again, please press star one to ask a question. We will now take our first question from Adrien de Saint-Hilaire. Please go ahead. Your line is open.
Yes. Good morning, everyone. I hope you can all hear me well, and I hope you're all fine. I've got a few questions, please, if that's okay. First of all, Arthur, to come back on your comment, you said that Publicis is capturing the shift towards digital and e-commerce, yet you mentioned Sapient is down about 2% organically. Can you help me understand, reconcile those two statements? Secondly, I can see that China and Asia-Pacific have slowed down between Q3 and Q2. Can you be a bit more specific around what happened there? It seems to be a bit peculiar given the improving GDP trends. Maybe a last question for Steve. Can you discuss your expectations for the Q4 advertising markets? You said Q2 was down 23%, Q3 down 10%.
What is your view for Q4, and what does that imply, therefore, for Publicis Groupe in Q4? Thank you so much.
Thank you, Adrian. I guess we have already taken a lot in Paris, so we're going to start with London. Steve, I guess, with question number three, and then I'll move to one and two.
Okay. Hi, Adrian. Good morning. Yeah, as you say, the situation is changing literally every day. When we spoke to you last time, we saw that we were predicting that Q2 was going to be down 23%, and we were predicting that Q3 would be approximately half of that rate of decline. Over the intervening quarter, we've actually seen the rate of decline that we've actually experienced. We haven't finalized the numbers, but we think that's slightly lesser level of decline, so somewhere less than 12%, but maybe around 10% decline for the quarter. You're asking about Q4, and as Arthur says, the situation is really uncertain. It's impossible to predict what's happening around COVID. Literally, just in the last few hours, you look at the announcement that's happened by President Macron in France, the situation in the U.K.
We have regional spikes throughout the world, and I guess most of you on this call are from Europe, so you can see regional spikes and lockdowns in many parts of north of the U.K. and Benelux. We have markets that are in shutdown, like Czech and Slovak and Moldova, etc. The situation is really unpredictable, which is why when we've seen this improving sequential numbers in our performance, it's really difficult to predict what's happening in Q4. We're very much in the hands of what's happening with COVID during the remaining quarter. The trend that we see at the moment was definitely looking to further improvement, improvement in Q3 in the overall advertising market with very different trends by market. That, we hope, is continuing in Q4.
As I said, just to reiterate my comments, it's a huge amount of unpredictability, as I'm sure all of you can understand. You saw the impact and the suddenness of the decline that happened in Q2 and Q3 as the pandemic swept across the world. So we are predicting at the moment better numbers, but it's very difficult to be objective due to the uncertainty.
If I may add just one point on that, is that we definitely are solid, again, in this context, positive trends within Publicis until year-end. But as Steve said, we have to be extremely cautious, honestly, only because of the spread of the virus for Q4. I think the only thing I will add that I made in my comment at the beginning, which is very important for us, I've been able, and Steve can confirm that, to meet many of our key clients and sometimes face to face, by the way, kind of like it, in the last two months with all the social distance, by the way. As I said, it's good to see their fighting spirit wherever they sit in terms of business journey.
This is pretty encouraging, and it gives me the opportunity to switch to your question one because at the end of the day, the question is not, "Will they spend less in the future?" It's, "How are they going to spend their investments?" The demonstration we are making today is that, of course, we see some figures on many clients and in many industries. But the fact that our clients are shifting now to more digital media, to commerce, to direct-to-consumer, where we have a strong offer with Epsilon, and you have seen the performance, and Sapient is helping us in a meaningful way to mitigate that.
By the way, we should not forget something that is important, is that there is the way we are able to grow organically, Sapient and Epsilon, but there is also the halo effect they are having on our clients to see us as a partner in their transformation, which is pretty important. To be more specific about Sapient, I will make just a few points. First, it is important to remind that Sapient in the U.S. in Q1 was actually positive. So we were off to a good year. You need to know that it's, of course, a business that is run by projects and that many of their projects have been stopped, or some of their projects have been stopped, but some very important clients that have been particularly affected by the crisis.
Not only do we see a recovery when you see that they are roughly in the U.S. at the same level as the rest of our business in the U.S. that is pretty resilient, but we see good momentum in the coming quarter. I want to be very cautious because if, for example, financial services that could have started to invest again stopped for other reasons, it's independent of our business trends. But we feel confident about what we are building with Sapient, and definitely the move we had done onto industry practice under the leadership of Nigel Vaz and being able to connect that with our clients is making a big difference.
If we had more time, I would tell you also, but that's a bit technical, but I'm sure we'll have a question on data later, why it is so important in a company like Publicis to have a good balance between paid media that you know very well, but also own channels, everything that has to do with commerce, with mobile app, because the balance between both is going to be important, and this is what, of course, Sapient is bringing. On China, I'm not going to come back on the number because Jean-Michel gave you most of the expectation. It's comparable or very high. Yes, the performance has slightly deteriorated this quarter, but the truth is that it is mainly reflecting the country's higher exposure to some industries that continue to be impacted by the crisis, like automotive.
When you go into this exercise that we did for the top 200 clients, you see that we have, to give you an idea, we are not heavyweights on luxury. We are heavyweighting on automotive. Now, I feel I shouldn't say that, but I feel pretty confident for the quarter to come. Just when I look at our new business pipeline, we won very recently McDonald's, Nestlé for all the production, TikTok, which is an important win there because it shows how much advanced we are on a data-driven approach, also in China, FCA, many new businesses that make us feel that our offer is at the right place. By the way, as you know, in terms of relative value, China is not that big for us, but it's a very important market for many of our clients.
The most important thing there for us is to have a competitive offer, and we are definitely making the demonstration with new business.
Thank you very much, all. Stay safe.
We will now take your next question from Lisa Yang from Goldman Sachs. Please go ahead.
Good morning. Thanks for taking my question. The first one is on your client mix. You talked about the top 200 clients accounting for seasonal revenue and being flat to slightly up in the first nine months, which I think implies the rest is probably down in the high teens. I'm just wondering if you could give us a bit more color in terms of what's happening with the long tail and whether you expect that to potentially improve in Q4. That's the first question. The second one is on the outlook, and I appreciate there's still a lot of uncertainty out there, and you mentioned all the potential new restrictions being announced. But your comments so far in the U.S. sound pretty good to me, at least.
I'm just wondering which markets do you see as potentially most at risk of a deterioration in Q4 because you said Q4 could be slightly worse than Q3. I'm just wondering what are the markets where you see a bit more risk at this point in time based on your conversations with clients. The third question is on Sapient, just to follow up to the previous question. I'm just wondering how much of the Q3 recovery was led by a catch-up effect there given all the projects were paused in Q2, or do you expect more catch-up to come in Q4? Given clients are looking to accelerate their digital transformation, are you seeing a lot more digital transformation pitches where Sapient is taking part at the moment? Maybe you can help quantify that as well. Thank you.
Thank you very much. Wow, that's a lot. Maybe I'll start with Sapient because I couldn't note everything, and I want to make sure that I answer that properly. What we see in Q3 with Sapient is what we experience in many of our places is that our clients had more activities in Q3 than they had in Q2. I don't think it is a catch-up. It's just, again, some projects coming back. We are hoping that this dynamic will continue. Again, I have to be extremely cautious, even more when I look at our president yesterday night. I don't want to tell you that there is a good pipe. Is this pipe will materially transform? It's too early to say. We are more in a slow recovery that can either accelerate or decelerate.
But when, again, I talk about a recovery, it's not our business trend. Again, remember that Sapient was positive in Q1, and we were off to a good year. But when our clients stop projects, the probability that they put it back on the table is massive because they will need to transform, but there is definitely a lag time. The second thing that is important in what you said, and by the way, this is how we're trying to work now moreover, is that the truth is you don't really pitch for transformation. You have to do work with your client, and this will lead me to the top 200 clients in a second, is you have to be on their side, will it be because you have a media relation, a creative relation, or even with Epsilon.
Actually, the synergy between Epsilon and Sapient is very, very interesting in terms of offer. Instead of pitching, you go as a consultant. You start by evaluating the potential. You then produce a roadmap, and then you execute. So we are not pitching that much. By the way, when it happens, again, some of our competitors like Accenture, we're having a pretty good transformation rate, if not very good. But the truth is most of the projects are long-term projects with clients that we're partnering in their transformation. This leads me to point number two, which is your question about the top 200 clients. This is a very important indicator for us because it is definitely on the top 200 clients that we are putting our model first.
This is with their clients where we have scale, where normally we know them very well, starting with the CMO and sometimes the CEO, that we can gain their trust to do more than a normal holding company and bring our integrated model in a business-fashioned way, which is really helping them to drive their journey of accelerating in growth while reducing in costs. So again, our top 200 clients are actually working better than the rest, first of all for this reason, which is it is where our model has been mainly deployed to our largest clients. The second thing that is important, and we have to be fair here, is that the regional mix, the region, is of course very in favor of the US-based clients, which, as you have seen in our results, are having a better organic growth.
It's not the word, but a more resilient organic growth than the rest of the world. Then it's important to note also, and this is why we wanted to show you for the first time by category, is that we have a higher exposure in our top 200 clients to resilient industries like healthcare and like FMCG. Industries that have been hit harder by the crisis, like leisure, are less represented into the top 200. But the truth is, and it's a reality, I guess, for everyone, and you see it, by the way, between the big agencies and the small, and I'm not talking about Publicis. I'm talking about the market here, is that generally, smaller clients have been suffering more in this crisis than the larger. I think I touched on all your questions, am I right?
You had a question also about the outlook that I see here. So it's a long answer, but hopefully, I'm touching many of the points or many of the questions that you could have. I don't want to come into one country particular to the other for a very simple reason. It is the main and almost only variable in Q4, knowing that we have very strong—no, very strong—strong business trend in the context. The main factor is the pandemic and is the one that will make the difference. Telling you how the pandemic will evolve, it's impossible at the moment, although what we have experienced in the last two or three days has actually accelerated our consciousness. By the way, it's important to note that in this context, Q4 is traditionally an adjustment period for advertisers.
They will make the right arbitrage depending on how things go in the coming weeks. By the way, they will make arbitrage in things where we can suffer particularly, and that are very present in this period, like events, for example. For all of these reasons, and I'm not going to go into the presidential election in the US or any other event like that, there are many reasons why we want to be cautious. You can count on us, as we did for the last three quarters, to do everything we can to continue to help our clients transform, to mitigate the impact of the revenue cuts. On the other side, as Jean-Michel said, manage our costs and our cash on a daily basis to make sure that we deliver on the objective we set.
Thanks, Arthur. That was really helpful. I'm just wondering if I can quick follow-up. You said the model has been mainly deployed amongst the top clients, which sort of clearly makes sense, and that's where you had most attrition in the past. But I mean, if you look at the long-term trends, it looks like the smaller clients have been outperforming the top clients. I mean, if you just look at the performance of Google and Facebook in general. So how long do you think it's going to take you to also roll out that model to maybe the long tail? How far are you in that sort of migration process?
You're actually touching a very, very interesting topic, which is, first of all, we don't have the same metrics as Facebook on what we call small clients. For them, small clients are SMEs, roughly, which are clients that we don't work that much with. It's small clients that are putting most of all of their investment with a platform like Facebook and Google in order to deliver personalization at scale. We don't have the same indicators for that, and this is very important. Once I've said that, what is true is that there is still potential for those SME clients. This is roughly if you take the U.S. clients that have revenue that are below $1 billion.
As you might have seen, we launched an offer in the middle of the crisis called The Pact that was focusing on what Facebook and Google call smaller clients, where we have been able, thanks to the accuracy of our data due to the lockdown, to guarantee outcome. It is only starting, but we see a potential for growth in an area that is honestly pretty new to us, not to Epsilon and not to Conversant in particular, but where there will be potential. So if I've got to resegment taking your question, there are our top 200 clients that are significant in terms of size. They are what we call the smaller clients that are very big clients for Facebook and Google because they have a CMO, they have an organization, they have a multimedia plan, and by the way, they are building a system to transform.
And they are the smaller, the SMEs, where we might have opportunities in the future if we continue to build our offer as we are.
Great. Thank you very much.
We will now take our next question from Tom Singlehurst from Citi. Please go ahead.
Bonjour, Arthur. Jean-Michel, thanks for taking the question. A couple of very quick ones, both on exposure by sort of client type. First one, FMCG, you mentioned it's sort of broadly stable. There's been some commentary from individual clients that they're looking at expanding budgets going forward. I suppose the question is, is this a sort of permanent change of direction, do you think, within FMCG? Are we past the hump whereby FMCG is now a tailwind rather than a headwind? That was the first question. The second one, very briefly, feels like Epsilon is very reliant on autos. So it makes that performance in the third quarter especially impressive. I'm just wondering whether that was because of autos or despite autos. Then the link to that, how you sort of diversify the sort of exposure at Epsilon going forward. Thank you.
Bonjour, Tom. Thank you for that. I'm making sure that I don't forget anything, so I'm noting everything. For me, it's always difficult as an account guy. It's very difficult to talk about FMCG because it is a mixed bag, honestly. You have clients that are doing great and particularly well in this period, and you have others that are still struggling. I will say the difference between the ones that are succeeding and the ones that are having difficulties at this stage is how they have entered the crisis. By the way, it's true for almost every industry, but if they entered with a strong strategy, with good products, with an innovation investment, and investing in their brand overall, you can see that some of them are doing brilliantly. By the way, it comes back to the segmentation I was making earlier.
I think it is the first time in my life as running an agency since 20 years that I'm having meetings with clients that have been performing so well in the last quarter that now their obsession is, "Okay, we are an FMCG company. We have seen the kind of growth that we haven't seen in a decade. How are we going to be able to maintain this growth in the future?" That's particularly interesting for us because it comes back to first-party data. It comes back to their ability to treat those new clients at an individual level and activate new products and offers to keep the kind of sales they're having. So it is a mixed bag.
In this mixed bag, I would say that the reason why we are doing well is mainly because we had a new business trend that is definitely supporting this kind of number. To make a long story short, we have some FMCG clients that are doing very well, and we are growing with them. We still have that are in difficulties, and we are bearing the pain with them, and we are here to help.
But overall, our new business momentum has helped us in this category. Your question on Epsilon is absolutely key, and it gives me the opportunity to tell you a few things while I'm answering your question. First, it's very important to note that Q3 has improved significantly versus Q2 for Epsilon. Bear with me a second because you're going to understand how it splits and how we can explain those numbers.
Via Auto, but also the retail practice, which are mainly non-food retailers, by the way, represent 40% of their revenue. It has been declining year to date, double digits, for the reason you know. If you are a non-food retailer, mainly a small one, you're closed. Auto, I'm not going to come back on that. You know it. So 40% of their business on auto and retail practice has been declining double digits. But of course, what is very encouraging there is that the other 60% of the revenue, at the opposite, has been largely positive. So it makes a demonstration that, of course, we are not immune to industry impact, for sure. When a car manufacturer closes their usines, it's by definition creating a problem that we have to manage in terms of funnel.
At the end of the day, I think it's very interesting to see it like this, and it explains the contribution of Epsilon to the group growth. As you know, it was flat in the US, and it's making a difference in our overall result in the US. I would insist just on one thing there, and then I promise we pass to the next question. It's something that honestly strikes me and makes me confident for the future, is that one thing is to track Epsilon organic growth, and of course, we are, and we want this business to deliver growth. What impressed me the most at the moment is the halo effect it's having on our other business, and by the way, a very good reason why we have won what we have won in new business again.
At the moment where we are going into a third cookie-less world, our ability to come with data that are real identities, real people with their name, with their address, with their email, with their mobile number, that we can follow through 7,000 attributes, looking at more than half of what they buy with a credit card, this is a massive asset for the client that wants to start to build direct relationships with their clients. This halo effect goes beyond the result we can have with Epsilon.
That's very clear.
Merci Tom.
Thank you very much. Merci.
We will now take our next question from Conor O'Shea from Kepler Cheuvreux. Please go ahead, sir.
Yes, thank you. Good morning, everybody. Thanks for taking my questions. Three, hopefully, quick questions for me as well. Firstly, Arthur on Sapient, just wanting to understand why the phasing of projects is so different in Q3 for the U.S. business, relatively positive versus some of the European countries that you called out where we're down a double digit. Second question, in terms of the top 200 clients, you mentioned finance and autos down, I think, significantly. That's, I think, a third of your client revenues group-wide. Can you give us a sense of how much that was down for the top 200 clients, those two sectors? And then the last question, maybe for Arthur or Jean-Michel, on margins, you called out that you're trading ahead of forecasts for the full-year margins.
Could that send a sign to your staff that they could increase their expectations for incentives at the end of the year? Is that a potential risk? And linked to that, can you give us a sense of how much the permanent headcount has reduced since January? Thank you.
I'll go fast on Sapient. It is a phasing issue. The fact that what you need to get there, and I guess you have experienced it around you, is that end of March, every project that could have been stopped has been stopped brutally for obvious reasons. Then the question is how fast the confidence of our clients is coming back to start building a pipeline and investing again in business transformation. It's true that the U.S. went faster than the rest of the world. As simple as that. They have been more bullish into that, and they are definitely that. I have to admit that I didn't get exactly your question on the top 200. Could you just what's your information?
Yeah, just for the finance and the auto sector, I think you said that you grew at 0.8% overall despite a decline for those two sectors. Can you give us a sense how much those two are down?
Yeah. So finance year to date is down at minus 8.2%. It's important to understand that here there is a mix that is mainly Sapient, by the way, because we are talking about big transformation projects that have been stopped for the moment. At the opposite, on auto at minus 9%, it is across the board because, of course, for several weeks, they have stopped advertising because they didn't have anything to sell in their dealership. That's roughly this. Jean-Michel?
Regarding the margins, we have this cost reduction program which has been launched, Conor, as you know, that we will deliver the $500 million. You probably noticed that already at the end of H1, we deliver almost $300 million. So we are confident that we will realize that, but it will be a lot depending on the level of revenue with the uncertainty of Q4. You understand that. So what is good is that in that difficult context, we have demonstrated the variability of our cost base. This is very important. Of course, the good thing is that if growth is back, we hope to be able to convert with this OpEx base having been reduced. We hope that we'll be able to convert the new revenues at a very good rate to deliver a good margin, of course.
In terms of staffing, we have not communicated on that, but we have, of course, adapted the level of staffing to the level of revenue. We are rather lower than the level of deterioration of the growth rate, but not very far, to be clear. Regarding the incentive, of course, this year, with the type of work that people have done, staying at home and so on, in very difficult contexts, this is not the year where we will reduce the bonuses and so on. So incentive will be at the right level this year to reward the work which has been done by our people.
No, I think that just to follow up on that, and it's true that the effort that has been done by our team in this period is absolutely outstanding, and we shouldn't undermine the difficulties that are creating the fact of working from home. There are good things about that. You can go through more transactional relationships, but at the end of the day, it is something that is difficult to manage, challenging, and we are trying to find the right balance now.
Yeah, yeah. Many thanks. Thank you.
Thank you. We will now take your next question from Matthew Walker from Credit Suisse.
Thanks very much, and good morning, everybody. I've just got a couple of financial questions. The first one is just in terms of the shape of the different years. If you think about it, as you improve your growth this year, or the rate of decline is less this year, that creates a slightly harder comparison for next year. Do you think that growth that might be expected next year might come in a bit lower because the comparison is tougher? Around Q1 for next year, to that theme, do you expect a positive Q1 because they had a pretty tough comparison from last year? Or do you expect a negative Q1 to continue this trend? The other question is on the costs.
How much of the $500 million comes back into the cost base for next year? And what about the sort of normal sort of wage inflation? Should we be thinking $200 million of the $500 million comes back in next year plus the wage inflation, or will the wage inflation be included in the $200 million? That's the sort of question I'm trying to ask.
Thank you. I'm going to take the first one pretty fast, which is, honestly, when it comes to 2021, it is far too early to give any kind of guidance at this stage. When you take into account the current uncertainty, only linked to the virus, to be clear. As you know, we had a very good start of the year. We were feeling confident. Hopefully, today, we are making the demonstration that we have a very strong foundation thanks to our offer, thanks to our model to weather the crisis. But it's far too early given the current uncertainties that are linked, again, to the virus, to give you more color or more flavor at the moment. Jean-Michel, on the $500 million?
The first point to answer Matthew's question is we are not planning for wage inflation and so on. The way we manage the personnel cost, we manage that relative to the level of revenue. Everything that we do in that respect is relative to the level of revenue. This is very important to understand that. This means that we want to protect a key ratio for the group, which is fixed personnel cost plus freelance. This is in the kitchen a little bit that I'm bringing you a little bit with that. But it is very important to understand that. So no projection of wage inflation as people have to budget for next year, understanding this fixed personnel cost and freelance ratio relative to the level of net revenue. This is the first point.
The second point, of course, in the EUR 500 million of cost reduction, a big part is on the personnel cost, but it is due to the downsizing in respect of the level of revenue. This means that if the revenues are back, this reduction that you will see in 2020 will partly disappear. We will keep only a part of this personnel cost saving coming from structural measures that we have implemented in 2020, and this will benefit the future, for sure.
But in the EUR 500 million, a big part, more than half, is on the G&A part of the operating G&A part of our P&L, which a part will be kept because we don't travel. For instance, we don't think that people will go back to travel next year as it was in the past, for sure. So a significant part will be still saved in 2021.
But we have also another element, which is the effect of the real estate consolidation, that we will have still a new wave in 2021 that, of course, we are managing properly to crystallize that in our P&L for next year.
Jean-Michel, I promised that we will be done at 10:30. It is 10:30, so maybe we want to take a last question if there is one more because we want to make sure that we leave you on time. And of course, Alessandra will be there to take every question offline. Is there one more, or are we done?
Thank you. We will now take our next question from Julien Roch from Barclays. Your line is open. Please go ahead.
Yes. Good morning, everybody. The first question is on net debt. You gave the numbers at the first half in the nine months. So net debt went down by $40 million in Q3. We don't really know what the free cash was last year because net debt went up a lot through Epsilon acquisition. So anything you can tell us on free cash in Q3, but also your expectation for Q4 and the full year? That's my first question. Then the second question is, thank you very much for the new revenue disclosure with indication on growth of creative and media and the rest. That's fantastic. But could we have the same split done for the U.S. for the group? So in the U.S., you're 25% creative, 30% media, 20% Epsilon, 15% Sapient, 10% health. Can we have that for the group?
I guess the two questions are for Jean-Michel.
I will answer with the second one because this is fresh in my memory, fresher in my memory. We are not managing that way anymore. This is very important to stick to the country model, which is extremely important for us. Everything is coming from that now in terms of managing the operation. There is no more media solution, no more Publicis Communications at the worldwide level. This is based on country, and which is allowing us to do a few changes and few transfers, few simplifications within the countries, which implies that it will be not very significant to give you creative worldwide or media worldwide or Sapient worldwide anymore. This is for the second question. Now, on the net debt, you have to—I saw your comment that you write this morning, for sure. I'm following what you are saying.
There is something that you have to take into consideration, that the comparison with Q2, Q3, sorry, 2019, which is extremely important. We have reduced significantly the level of debt versus end of September 2019. We have also to consider the following element for this year. We have a lower dividend paid in cash in Q3, that you know, for all the reasons that you know. You have a second element also, no M&A in Q3.
We have a third element also, which gives you some explanation regarding the net debt at the end of September. It is an FX impact on our net debt, which is mainly in dollars, as you know. Of course, we will continue to monitor our cash situation on a day-to-day basis. We expect our net debt at the end of 2020 to be slightly below end of 2019.
Thank you, Jean-Michel. Look, I want to thank you again for listening this morning and asking the question. I'm sorry we're a bit longer than expected. As I said, of course, Alessandra and Boris are here to take all of your questions offline. I mean, hopefully, what you took out of this call is that we have the offer that our clients need and the structure to perform financially to make sure that we can weather this crisis. This is the main message we have for you. We will be, again, very happy to give more color offline. I'm just going to hope that everyone is fine, that you can take good care of you and your family in this period, and see you very soon. Thank you very much.
Thank you. That will conclude today's conference call. Thank you for your participation. Ladies and gentlemen, you may now disconnect.